Trials of rhDNase in mixed groups of adults and children with cystic fibrosis (CF) have demonstrated improvements in lung function and well-being. This has led to many pediatric CF patients receiving regular rhDNase therapy although their response to treatment may not be the same as that seen in adults. We have retrospectively reviewed the effects of rhDNase during the first year of therapy in 65 children receiving the drug at two tertiary referral centers. Outcome measures included changes in lung function, oxygen saturation, use of intravenous antibiotics, and subjective improvement. Median baseline lung function (% of predicted) was 45% for forced expiratory volume in 1 second (FEV1) and 58% for forced vital capacity (FVC). At 3-4 months following initiation of therapy the group demonstrated median (95% CI) increases of 14.2% (95% CI 7.3; 21.1%) in FEV1 and 7% (0; 14%) in FVC. Within this wide scatter of responses, one-quarter of children deteriorated, but almost 50% showed significant improvements of > 10%. A similar pattern was seen at 9 months, with median increases for the group of 11.1% (0; 18.8%) in FEV1 and 5.6% (0; 17%) in FVC, again with approximately one-third of the group deteriorating and one-half improving significantly. Intravenous antibiotic use decreased significantly. Almost all the children (89%), including those with a fall in lung function, described subjective improvement. There were no predictive markers at baseline of a good response to the drug. However, there was a good correlation between lung function response at 3 months and that at 6, 9, and 12 months. Thus, children respond to rhDNase at least as well as adults, and a therapeutic trial is justified in those over 5 years with significantly impaired lung function. Response is highly variable, making careful individual assessment mandatory. Baseline characteristics are not useful in predicting those children who will respond well to treatment, but long-term response to the drug can be predicted on the basis of spirometric improvement at 3 months. Therefore, this would be a useful time period for a therapeutic trial.